Q: what is the key motivation behind BAMP?
BAMP grew out of an area management project in the Broughton Archipelago intended to significantly reduce the likelihood that sea lice from farmed salmon would infect out-migrating juvenile wild salmon in the area. BAMP is intended to measure lice on farmed and wild fish and to assess the effectiveness of farm management actions.
Q: how are BAMP activities managed?
BAMP research activities are guided by a Science Team made up of a number of independent researchers. The broader operational governance of the program is the responsibility of these researchers working in tandem with representatives of the sponsoring organisations. In 2010 the BAMP Science Team established an initial set of research and monitoring protocols to support focused efforts to improve understanding of the interactions between farmed salmon, wild salmon and sea lice in the Broughton Archipelago. Both the research plan and the monitoring protocols are subject to annual review by the Science Team.
Q: what are sea lice?
Sea lice are small marine parasites that occur naturally on many different species of wild fish including wild adult salmon. The early development stages of sea lice are planktonic and are transported on water currents and tides. When they encounter marine fish they attach themselves, usually on the skin, fins and/or gills and feed off the mucous or skin. There are 13 known species of sea lice in the marine waters of British Columbia, but the common ‘salmon louse’ is the one we hear the most about. The Latin name for this salmon louse is Lepeophtheirus salmonis. In the spring, when fish eggs hatch and juvenile salmon migrate down the rivers and enter the ocean they may become infected with sea lice from salmon farms where they are located along wild salmon migration routes.
Q: how many wild fish are sampled and from which sites?
Since 2010 the BAMP field team has been sampling wild Pacific salmon at around 100 sites across the Broughton Archipelago to measure sea lice levels. At each site up to 30 fish of the species of interest (pink, chum and coho) can be retained if they are present. Once the data from 2011 are available for analysis an assessment will be made as to whether it might be possible to reduce the number of sites from which fish are sampled and if so, which sites.
Q: what are the typical summary measures used to report lice levels on fish?
There are a number of standard terms that will typically be found in descriptions of the level of parasites on fish, each of which helps illustrate different aspects of any infestation. The definition of some of the terms makes reference to the table below.
Sample Results (based on Trip 2 in 2011)
|Number examined||Average length
|Pink||L. salmonis||2.7% [1.7 - 3.7]||1.1 [1-2]||0.03||C. clemensi||9.2% [7.4 - 11.0]||1.2 [1-3]||0.11||Chum||L. salmonis||6.7% [4.7 - 8.8]||1.2 [1-2]||0.08|
|C. clemensi||14.7% [11.8 - 17.6]||1.5 [1-5]||0.22|
This measure simply describes the mean number of lice across all of the fish which were sampled - in this sense it is the closest to what we colloquially think of when we say, "the average". In the example above, for pink salmon we found a total of 28 L. salmonis lice on the 977 fish observed, and thus the abundance is simply 28 divided by 977, here rounded to 0.03 (the actual value is closer to 0.02866).
One of the problems with a traditional 'average' is that it is of limited use when there are many individuals who return a value of zero - in our example there are in fact 951 fish on which there is no infestation. The "intensity" measure also provides the mean number of lice, but only on those fish that are infested. In the example above, for pink salmon the 28 L. salmonis lice were found on 26 of the observed fish, and thus the intensity is simply 28 divided by 26, which we have rounded to the value of 1.1. We also report the range, the larger number in which indicates the maximum number of lice found on any one fish - in the case we have been looking at 2, or in the case of C. clemensi lice on chum salmon 5. [The thoughtful reader may observe that if every fish was found to be infested then the intensity and abundance would be identical, however, as can be seen from the data reported here this rarely the case for lice on wild juvenile salmon.]
In cases where overall abundance is low and many fish have no infestation it is often informative to report the "prevalence" which indicates the proportion of fish on which there was at least one louse present. In the case of L. salmonis shown above, 26 of the pink salmon observed had at least one louse present out of the total of 977 fish examined. Dividing 26 by 977 gives us a value of 0.0266 which, after rounding, can be expressed as 2.7% prevalence.
95% Confidence Interval (CI)
Each of the three measures noted above are referred to as "point estimates" for the values estimated to be representative for the population of fish from which these samples were taken. There will of course be variability associated with these values and it is often useful to summarise the range of possible values that might be likely to arise. There are a number of ways to express this variability but all of these (apart from the simple "range" shown in the case of intensity) are reasonably complex from the point of view of statistical interpretation. We have chosen to show the "95% CI" associated with prevalence values which for the 2.7% result for L. salmonis on pink salmon is expressed as [1.7 - 3.7]. This essentially indicates that the 'true' percentage of pink salmon that are infested with L. salmonis is highly likely to lie somewhere in the range between 1.7% and 3.7%. In general, the smaller the confidence interval the more likely it is that the point estimate - in our case for prevalence of lice - is close to the true value. [For those who have an interest in such things, the technically correct interpretation of a 95% CI is that if we were to go out and sample from pink salmon populations 100 times, then in 95 of these cases the true value would lie within the specified confidence intervals.]
Q: why does the BAMP protocol not simply count lice on live wild fish?
Since 2011 the protocol for assessing sea lice loads on wild Pacific salmonids has involved lethal sampling in the field, followed by a detailed analysis of lice and fish at one of two laboratories. Over the years alternative approaches have been used. Some researchers have used ‘live licing’ to assess the sea lice loads on wild salmon. Indeed, in 2010 the BAMP monitoring program used both live and lethal sampling to assess lice loads to gain an understanding of 'infection pressure'. As of 2011 the Science Team moved to ensure the highest quality and consistency of lice monitoring on wild salmon by assessing samples in the laboratory. This becomes particularly important when prevalence levels are very low – for example, on one trip at the start of the 2011 season only 8 fish from over 600 samples were found to be infected. In addition, when lice are in their early stages of development, as they often are on small fish, it is almost impossible to distinguish between the species of lice involved without using a microscope in the laboratory. Knowing whether the infestation is by L. salmonis as opposed to one of the Caligus species is an important factor in clearly understanding the interactions between lice on wild and farmed fish.
Q: what mechanism(s) are available within the BAMP research process to alter details of the field and/or laboratory protocols?
Each year the BAMP Science Team meets to review activities over the preceding year and to plan for both the research and field activities in the succeeding year. This is the main point at which alterations to overall plans and protocols are made. A key reason for this is that a face-to-face meeting is a more conducive environment to consider complex arguments and ensure that all viewpoints are adequately explored. However it is also seen as important that once the sampling season has begun the protocols remain consistent for the rest of that year. Of course if critical issues are raised in the course of carrying out project activities the Science Team can address these specifically in the interim. For example, in 2010 it was brought to the attention of the BAMP Science Team that the process of euthanizing sampled fish did not fully conform to welfare standards that many would consider essential. This important issue was dealt with immediately and a modified protocol adopted from the succeeding field trip forward.